add torch

Dockerfile

@@ -10,5 +10,6 @@ COPY ./requirements.txt /code/requirements.txt
 RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
 
 COPY . .
+RUN mkdir ./.cache/huggingface
 RUN chmod 777 -R ./*
 CMD ["python", "run_server.py",  "--llm", "Qwen/Qwen-1_8B-Chat", "--model_server", "http://127.0.0.1:7905/v1", "--server_host", "0.0.0.0", "--workstation_port", "7860"]

run_server.py

@@ -10,7 +10,7 @@ from pathlib import Path
 from qwen_agent.log import logger
 from qwen_agent.utils.utils import get_local_ip
 from qwen_server.schema import GlobalConfig
-
+os.environ["TRANSFORMERS_CACHE"] = "./.cache/huggingface/"
 
 def parse_args():
     parser = argparse.ArgumentParser()